Biochemistry 1991, 30, 3247-3255

Thorpe, P. E. (1985) in Monoclonal Antibodies '84: Biological and Clinical Applications (Pinchera, A,, Doria, G., Dammacco, F., & Bargellesi, A., Eds.) pp 475-506, Editrice Kurtis, Milano, Italy. Thorpe, P. E., Ross, W. C. J., Brown, A. N. F., Myers, C. D., Cumber, A. J., Foxwell, B. M. J., & Forrester, J. T. (1984) Eur. J. Biochem. 140, 63-71. Townsend, R. R., Hardy, M. R., Wong, T. C., & Lee, Y. C. (1986) Biochemistry 25, 5716-5725. Tressel, P., & Kosman, D. J. (1980) Biochem. Biophys. Res. Commun. 92, 781-786. Vancurova, D., Ticha, M., & Kocourek, J. (1976) Biochim. Biophys. Acta 453, 301-310. Villafranca, J. E., & Robertus, J. D. (1981) J . Biol. Chem. 256, 554-556. Vitetta, E. S . (1986) J. Immunol. 136, 1880-1887.

3247

Vitetta, E. S.,& Uhr, J. W. (1985) Annu. Rev. Immunol. 3, 197-212. Vitetta, E. S., Fulton, R. J., May, R. D., Till, M., & Uhr, J. W. (1987) Science 238, 1098-1104. Warren, L. (1959) J . Biol. Chem. 234, 1971-1975. Wei, C. H., & Koh, C. K. (1978) J. Mol. Biol. 123,707-71 1. Yet, M.-G., Chin, C. C. Q., & Wold, F. (1988) J. Biol. Chem. 263, 111-1 17. Yoshitake, S., Funatsu, G., & Funatsu, M. (1978) Agric. Biol. Chem. 42, 1267-1274. Youle, R. J., & Colombatti, M. (1987) J . Biol. Chem. 262, 4676-4682. Youle, R. J., Murray, G. L., & Neville, D. M., Jr. (1981) Cell (Cambridge, Mass.) 23, 551-559. Zentz, C., Frfnoy, J.-P., & Bourrillon, R. (1978) Biochim. Biophys. Acta 536, 18-26.

Cloning and Expression of Complementary DNAs for Multiple Members of the Human Cytochrome P450IIC Subfamily?,$ Marjorie Romkes,f Michael B. Faletto,$ Joyce A. Blaisdel1,s Judy L. Raucy,ll and Joyce A. Goldstein**l Laboratory of Biochemical Risk Analysis, National Institute of Environmental Health Sciences, Research Triangle Park, North Carolina 27709, and College of Pharmacy and Toxicology Program, University of New Mexico, Albuquerque, New Mexico 871 31 Received October 15, 1990; Revised Manuscript Received December 27, 1990 ABSTRACT: The present study characterizes the profile of cDNAs from the human P450IIC subfamily in

a library from one individual, and it describes three new members of this subfamily (IIC17, IIC18, and IIC19) isolated from two human cDNA libraries. cDNA libraries were constructed from two human livers which differed phenotypically in the hepatic content of P450 HLx (IIC8). The library from the phenotypically low HLx individual was screened by using a cDNA for rat liver P450IIC13 and an oligonucleotide probe for human IIC8. One clone, 254c, was isolated which clearly represents a new member of the human P450IIC subfamily (IIC17). This clone lacked the first 358 nucleotides at the N-terminus but was only 91% homologous in its nucleic acid sequence to IIC9 and 79% homologous to IIC8. Near-full-length clones for IIC9 were also isolated from this library, but no clones for IIC8 were found. Northern blots indicated that the mRNA for IIC8 was low or absent in this individual. A second cDNA library (from a liver phenotypically high in HLx) was then screened. Eighty-three essentially full-length (>1.8 kb) clones belonging to the IIC subfamily were isolated from this library. These include full-length clones for two additional new members of the IIC subfamily. Clones 29c and 6b appear to be allelic variants (IIC18), differing by one nucleotide (one amino acid change) in the coding region. Clone 1 l a represents a full-length clone for a third new P450 (IIC19). Both IIC18 and IIC19 are most homologous to IIC17 (87% and 95%, respectively). Full-length clones were isolated for two allelic variants of IIC9 (clones 25 and 65) which differed by three nucleotides, resulting in one amino acid difference. Hybridization analysis and partial sequencing indicated that of the 83 clones in this library, 50 were 1 of the 2 allelic variants of IIC9,29 were IIC8, 3 were the 2 allelic variants of IIC18, and 1 was IIC19. cDNAs for IIC9, IIC18, and IIC19 were expressed in COS-1 cells. Antibody to IIC9 recognized the proteins in cells transfected with cDNAs for IIC9; however, neither antibody to IIC9 or IIC8 recognized the other two cytochromes, suggesting that these cytochromes differ immunochemically from IIC8 and IIC9. COS-1 cells transfected with IIC18 showed an increase in metabolism of mephenytoin, but no metabolism of mephenytoin was detected in COS-1 cells transfected with IIC9 or IIC19. C y t o c h r o m e P450' enzymes are important in the oxidative metabolism of both endogenous substrates and xenobiotics. Genetic polymorphisms of P450 enzymes result in distinct subpopulations which differ in their ability to perform par+ A preliminary report of a portion of this work was presented at the 81st Annual Meeting of the American Society for Biochemistry and Molecular Biology (Romkes et al., 1990). *Thenucleic acid sequence in this paper has been submitted to GenBank under Accession Number 505326. * To whom correspondence should be addressed. *National Institute of Environmental Health Sciences. 11 University of New Mexico.

ticular drug biotransformation reactions. We have recently found that a polymorphism in rat P450g (IIC13) in outbred rats is due to the presence of a few single base mutations in the mRNA resulting in simple amino acid changes (Yeowell et al., 1990). Human P450 HLx, which is immunochemically We have used the recommended P450 gene nomenclature system (Nebert et al., 1991) throughout this paper. All new P450IIC sequences were submitted to Dr. Nebert for inclusion in the 1991 update to this nomenclature system, and the numbering system is consistent with this update. Abbreviations: SSC, standard sodium citrate; SDS, sodium dodecyl sulfate.

This article not subject to US.Copyright. Published 1991 by the American Chemical Society

3248 Biochemistry, Vol. 30, No. 13, 1991 related to rat P450g (IIC13), is phenotypically variable in humans as determined by immunoblot analysis (Wrighton et al., 1987). Okino et al. (1987) isolated a cDNA (IIC8) which appears to encode the protein for human HLx, on the basis of its N-terminal sequence. Other research groups have isolated cDNAs MPl2/MP20 (Ged et al., 1988) and IIC2 (Kimura et al., 1987) which are presumably allelic variants of IIC8. However, the basis of the polymorphism has not yet been explained. At least three or more forms of P450 belonging to the human TIC subfamily have been isolated by different laboratories, including three proteins termed P450Mp-1, P450Mp-2, and P 4 5 0 ~ (Shimada ~.~ et al., 1986; Lasker et al., 1987; Kawano et al., 1987; Gut et al., 1986; Beaune et al., 1985; Ged et al., 1988). There are also other polymorphisms in this subfamily. For example, there is a polymorphism in the metabolism of the S enantiomer of mephenytoin in humans. Guengerich and co-workers (Shimada et al., 1986) have isolated two forms of P450 (MP-I and MP-2) which appear similar in their (S)-mephenytoin hydroxylase activity. They then isolated the cDNAs for the highly related MP-8 (IIClO) and MP-4 (IIC9) (which varied by only two bases in the coding region) using antibody to P45OMP.,(Umbenhauer et al., 1987; Ged et al., 1988). These may represent allelic variants of the same protein but have seen classed at present as separate genes on the basis of differences in the 3’-noncoding region. Several other groups have also isolated cDNAs which appear to be allelic variants of IIC9 (Kimura et al., 1987; Shephard et al., 1989; Yasumori et al., 1987). However, the majority of studies from a number of laboratories using different cDNA expression systems indicate that various IIC9 variants do not appear to metabolize (S)-mephenytoin. For example, when the cDNA isolated by Kimura et al. (1987) was expressed in HepG2 cells, it metabolized racemic mephenytoin but had no (S)-mephenytoin hydroxylase activity, suggesting that some other isozyme is responsible for the polymorphism in the metabolism of (S)-mephenytoin. Similarly, when a complete MP-8 clone (constructed by using the five N-terminal amino acids determined from amino acid sequencing of MP-1) was expressed in yeast, it metabolized tolbutamide but did not 4-hydroxylate (S)-mephenytoin (Brian et al., 1989). However, Yasumori et al. (1987) isolated a variant of llC9 differing by only two amino acids from MP-8 (positions 358 and 41 7) and by only three amino acids (positions 4, 6, and 144) from the IIC9 cDNA isolated by Kimura et al. (1987). His laboratory reported that the protein encoded by this cDNA metabolized (S)-mephenytoin when expressed in yeast (Yasumori et al., 1989). The role of the various allelic variants of IIC9 in (3)-mephenytoin metabolism remains controversial at this time. Genetic analysis has suggested the presence of 1 7 genelike sequences in the IIC subfamily. Therefore, it seems likely that this subfamily may include other members. In this study, we constructed two cDNA libraries from individuals phenotypically high and low in HLx, to examine whether a variant mRNA for IIC8 was responsible for the polymorphic expression of HLx and to identify additional members of the IIC subfamily. No clones for IIC8 were isolated from the individual phenotypically low in HLx, while several full-length cDNAs for IIC8 were isolated from the phenotypically high individual. Two allelic variants for I1C9 were isolated. In addition, a partial cDNA for a new member of the IIC subfamily (IIC17) and full-length cDNAs for two additional new members (IIC18 and IIC19) were isolated. These new members of the IIC subfamily were expressed in COS-I cells and

Romkes et al. shown to be immunochemically distinct from HLx and IIC9, and IIC 18 metabolized racemic mephenytoin. MATERIALS AND METHODS Materials. Human liver samples were obtained from organ donors through the National Disease Research Interchange in Philadelphia, PA, and from the Human Liver Research Facility, Stanford Research Institute, Life Sciences Division, Menlo Park, CA. Restriction endonucleases were purchased from Pharmacia LKB Biotechnology, Inc. (Piscataway, NJ). [ c Y - ~ ~ P ] ~(3000 C T P Ci/mmol) and [y-32P]ATP (5000 Ci/ mmol) and [ C Y - ~ ~ S I ~(650 A T PCi/mmol) were from Amersham Corp. (Arlington Heights, IL). All other reagents were of the highest quality available. Purijiied Proteins and Western Blots. Cytochromes P450IIC8, P450IIC9, and NADPH:P450 reductase were purified and antibodies to IIC8 and IIC9 prepared in rabbits as previously described (Lasker et al., 1987). Human hepatic microsomes were prepared from several different liver organ donor samples as previously described (Goldstein et al., 1982). The microsomal protein was determined by the method of Bradford (1976). SDS-polyacrylamide gel electrophoresis (7.6% acrylamide) of hepatic microsomes was performed according to Laemmli (1979), and the proteins were transferred to nitrocellulose. Western blots were performed as previously described (Yeowell et al., 1985) using polyclonal antibodies to IIC8 and IIC9 (Raucy et al., 1989). Construction and Screening of Human Liver cDNA Libraries. Total human liver RNA was prepared by the guanidine hydrochloride method (Cox, 1968) from two human livers either low (860624) or high (S33) in HLx as identified by immunoblot analysis. Poly(A+) RNA was then isolated by two passages over an oligo(dT)-cellulose column (Aviv et al., 1972). The low-HLx cDNA library was prepared by Stratagene Cloning Systems (La Jolla, CA), and the double-stranded cDNA was treated with S1 nuclease. Following the addition of EcoRI linkers, the double-stranded cDNA was size-fractionated on a CL-4B Sepharose column. The largest fraction was ligated into XZAPII and then transfected into XL1-Blue. The high-HLx cDNA library was constructed in our laboratory following the methods of Watson and Jackson (1985). Double-stranded cDNA was ligated to EcoRI linkers, size-fractionated on an agarose gel (1.8-2.4 kb), and then ligated into XZAPII (Stratagene) and transfected into XLlBlue. The low-HLx library was screened under conditions of low stringency with a 32P-labeledrat P450IIC13 cDNA probe and with oligonucleotides for human IIC8 (T300R) (5’-TTAGTAATTCTTTGAGATAT-3’) and IIC9 (M300R) (5’CTGTTAGCTCTTTCAGCCAG-3’). The high-HLx library was screened under conditions of low stringency using a 32Plabeled 254c (IIC17) cDNA probe and M300R (IIC9). Positive clones were isolated, transfected into XL1-Blue, and excised into the plasmid Bluescript, according to Stratagene’s excision protocol. Sequence Analysis. The Bluescript plasmids containing the positive cDNA inserts from the low-HLx library were purified by CsCl gradients, while the plasmids containing cDNA inserts from the high-HLx library were purified by using Qiagen plasmid purification kits (Qiagen, Inc., Studio City, CA). The double-stranded cDNA inserts were sequenced by the dideoxy chain termination method for Sanger et al. (1982) using Sequenase kits (U.S. Biochemical Corp., Cleveland, OH). The nearly full-length clones 186 and 254c and the full-length clones 65, 25, 7b, 35g, l l a , 29c, and 6b were sequenced completely in both directions with primers spaced approxi-

Cytochrome P450IIC Subfamily mately 200 bases apart. The remaining positive clones from the high-HLx cDNA library were sequenced in both directions through both the 5' and 3' ends and through all the regions which would identify any of the known allelic variants. Human RNA Blot Analysis and Hybridization Conditions. Poly(A+) RNA (10 pg) was electrophoresed in a 1% agarose gel under denaturing conditions and transferred to a Nytran filter (Micron Separation, Inc., Westboro, MA), and filters were then baked for 2 h at 80 "C. The filters were prehybridized for 2 h, then hybridized overnight with a 32P-labeled specific oligonucleotide probe for TIC8 (T300R) at 42 "C, washed 3 X 5 min at room temperature and 1 X 5 min at 42 "C with 2 X SSC/O.l% SDS, and radioautographed. Filters were then stripped with 5 mM Tris (pH 8.0), 0.2 mM EDTA, 0.05% sodium pyrophosphate, and 0.1 X Denhardt's for 2 h at 65 "C and rehybridized with a random-primed actin cDNA (Oncor, Gaithersburg, MD) at 50 "C using 6 X SSC, 4 X Denhardts, and 0.5% SDS. These filters were washed 1 X 5 min at room temperature, 1 X 10 min at 48 "C, and 4 X 15 min at 48 "C and radioautographed as before. The IIC8 mRNA band was quantitated by scanning with an LKB U1trascan laser densitometer, and the values of the integrated peaks were divided by those of the actin peaks. Hybridization and washing conditions for screening libraries with random-labeled cDNAs for IIC13 (g) or IIC17 used the same solutions as described for actin, but were performed at nonstringent temperatures (42 "C). Conditions for hybridization of clones with T300R were identical with those described above. Hybridization of cDNA clones with M300R (recognizes 1IC9, IIClO, and ITC19), 254c-351R (IIC17) (5'-ACTTTTCAATGTAAGCAAAT-3'), and M 1724R (I IC9) (5'-CATATTAATAACTCAACATT-3') was identical except that for each oligomer the hybridization temperature and the high-stringency wash were 5 "C below the calculated melting temperatures. COS-1 Cell Expression Studies. cDNA inserts were ligated into the cloning region of the expression plasmids pSVL (Pharmacia LKB Biotechnology, Inc., Piscataway, NJ) or pcD (Okayama & Berg, 1983) and used to transform COS-1 cells. COS-1 cells were plated at (1-2) X lo6 cells per IO-cm dish and grown for 24 h in Dulbecco's-modified Eagle's medium with 10% fetal bovine serum (DMEM). The cells were then washed with Dulbecco's phosphate-buffered saline (PBS) and transfected with recombinant plasmid (3 pg per dish) in DEAE-dextran (500 pg/mL) for 30 min-1 h at 37 "C. The transfected cells were then treated with chloroquine (52 pg/mL) in DMEM for 5 h (Luthman & Magnusson, 1983), washed with PBS, refed with DMEM, and incubated for 72 h prior to harvest. Typically, 15-20 dishes were transfected with each recombinant plasmid. For Western blot analysis of the recombinant transformed COS-1 cells, cells were scraped from the dishes into buffer (50 mM Tris-HCI, pH 7.5, 150 mM KCI, and 1 mM EDTA) and lysed with 3 X 5 s bursts with a polytron. A portion of each lysate was centrifuged at 9000g and then lOOOOg for the preparation of a microsomal fraction. Western blots were then performed as described above. Total RNA was isolated from transfected COS-1 cells, and Northern blots were performed as described for human samples. The filters were hybridized with a 32P-labeledoligonucleotide probe which hybridizes with all IIC clones isolated (IIC500R) (5'-GGAGCACAGCCCAGGATGAA-3') at 55 "C overnight, then washed 3 X 5 min at room temperature and 2 X 5 min at 37 "C, and radioautographed. Mephenytoin 4-Hydroxylation Assay. The methods for determining the enzymatic hydroxylation of mephenytoin were

Biochemistry, Vol. 30, No. 13, 1991

3249

anti-IC9

anti-llC8

FIGURE 1: Western blots of human liver microsomal proteins. Microsomal proteins were separated by SDS-polyacrylamide gel electrophoresis. Blot A was performed using polyclonal antibody to IIC9 and blot B with anti-IIC8 (HLx). Each lane represents 20 pg of microsomal protein from an individual liver. The IIC8 antibody also recognized purified rat P450IIC13 (g). cDNA libraries were constructed from livers 860624 (low HLx) and S33 (high HLx).

performed by using the general incubation conditions and HPLC assay previously described by Shimada et al. (1986). The microsomal fractions prepared from the COS- 1 cells, containing 200 pg of microsomal protein, 1000 units of purified P450 reductase, and 0.2 nmol of cytochrome b5, were incubated with 0.5 mM racemic mephenytoin and 50 mM potassium phosphate buffer (pH 7.4) in a final volume of 0.5 mL. The reactions were carried out at 37 OC, were initiated by the addition of 20 pL of 50 mM NADPH, and stopped 15 min later by the addition of 100 pL of 1 N HCl. The samples were extracted with water-saturated ethyl acetate and dried on low heat under nitrogen. The residue was reconstituted in 25 pL of 50% methanol, and 10 pL was injected onto the HPLC with UV detection at 204 nm using a Rainin 10-cm Microsorb C,* octyldecylsilyl "short one" column. The mobile phase was 55% methanol/45% H 2 0 with a flow rate of 1 mL/min. Statistical significance was assessed by a Kruskal-Wallis test for overall differences between groups, and pairwise comparisons were made with a Mann-Whitney U test. RESULTS Expression of cytochromes P450IIC8 and IIC9 was examined in human liver microsomes by Western blot analysis (Figure 1). Polyclonal antibodies to both IIC8 and IIC9 exhibited some cross-reactivity to related proteins. Both of these P450s are expressed constitutively; however, P450IIC8 (HLx) is phenotypically variable as described by Wrighton et al. (1987). The level of P450IIC9 did not vary appreciably in these tissues. Microsomes from liver 860624 contained IIC9 but were essentially lacking HLx, while liver microsomes from S33 contained high amounts of both proteins. Two cDNA libraries were constructed from human livers 860624 and S33, which expressed phenotypically low or high amounts of HLx, respectively. Screening the cDNA library constructed from a low-HLx individual with a cDNA for rat IIC 13 under nonstringent conditions and with oligonucleotide probes specific for IIC8 and IIC9 yielded several clones for IIC9 and a novel P450IIC cDNA. None of the clones iden-

3250 Biochemistry, Vol. 30, No. 13, 1991 Table I: Percent Homology for Nucleotide and Amino Acid Sequences of P45011C cDNAs" 254c 29c 1la clone IIC8 IIC9 (IICI7) (IICI8) (IIC19) 254c (llC17) 79 91 100 87 95 86 92 78 90 100 85 81 100 82 29c(11C18) 81 77 82 86 100 81 Ila (11'219) 81 90 95 82 IO0 79 92 92 81 100 'For each comparison, the upper value represents percent nucleotide homology, and the lower value represents percent amino acid homology. The nucleic acid comparisons include both the coding and noncoding regions. The IIC9 sequence used in this comparison was the cDNA sequence for human form 2 (Yasumori et al., 1987). Table 11: Distribution of P45011C cDNA Clones from Human Liver s33a no. of clones % distribution IIC8 29 35 IIC9 39 47 65 25 11 13 llCl0 0 0 IIC18 29c 1 1.2 6b 2 2.5 IIC19 ( I la) 1 1.2 total 83 100 'Clones were classified by hybridization with specific oligonucleotide probes and partial sequencing.

tified in this library were full-length. Clone 186 was identical with but 25 base pairs longer than MP-4, a IIC9 clone previously described by Ged et al. (1988). Clone 254c lacked 358 nucleotides at the N-terminus but clearly represents a new member of the IIC subfamily (IIC17), since it is only 90% homologous in amino acid sequence to IIC9 and 78% homologous to 1IC8 (Table I). The cDNA sequence for 254c is shown in Figure 2 (bottom line) along with the sequences of other cDNAs from the IIC subfamily sequenced later in this study. No clones for IIC8 were identified in this library, despite the screening of >90 000 plaques with a specific oligoprobe for IIC8. We then screened approximately 40 000 plaques from the library from liver S33 with the cDNA for 254c under nonstringent conditions and with an oligonucleotide probe specific for IIC9. Eighty-three essentially full-length IIC clones (>1.8 kb) were isolated, purified, and partially or completely sequenced (Table 11). Of these, 29 clones were found to encode cytochrome P450IIC8. We isolated two variant clones (7b and 35g) of IIC8 which were completely identical with Hpl-1 and Hpl-2 reported by Okino et al. (1987). Clone 35g is identical with 7b except that it contains an additional 39 bases of a presumed intron which results in the addition of two early stop codons. Both cDNAs were full-length, but 35g contained a longer 5' leader (81 vs 77 base pairs). The majority of the clones (50) isolated from the library from liver S33 coded for IIC9. Interestingly, all of the 50 clones appeared to be 1 of 2 IIC9 allelic variants, typified by the full-length clones 65 and 25. All of these clones were sequenced through the 5' and 3' ends and through regions which would identify known allelic variants. Thirty-nine of the IIC9 clones were identical with clone 65, and 11 were identical with clone 25. The nucleotide sequence for clone 65 is shown in Figure 2, and the differences for clone 25 are displayed in the line above (boxed). Clones 25 and 65 were identical in the 5'- and 3'-noncoding regions but contained

Romkes et al. three single-base changes at positions 842, 1075, and 1425. Two of these base changes were conservative, but the third would result in one amino acid difference at position 359 (isoleucine versus leucine). Figure 3 compares the amino acid sequences for the variants of IIC9 isolated by various laboratories. Clone 65 is identical in amino acid sequence with human form 2, although it differs by two silent changes in the coding region and four differences in the noncoding region (Yasumori et al., 1987). Clone 65 contained a leucine instead of a isoleucine at position 4, a valine instead of a serine at position 6, and an arginine instead of a cysteine at position 144 compared to the IIC9 sequenced by Kimura et al. (1987). The IIC9 reported by Meehan et al. (1988) has substitutions at positions 144, 175, and 238 compared to our clones. The remaining clones characterized from the human liver S33 cDNA library encode several novel P450IIC cDNAs. Their DNA sequences are shown in Figure 2. Two of these clones, 29c and 6b, differ by one nucleotide in the coding region (position 1154), which would result in a single amino acid change (threonine vs methionine at position 385). Clone 29c had a very long (198 bp) 5'-noncoding sequence and a polyadenylation signal 21 bases from the poly(A) tail. Clone 6b did not have a polyadenylation signal at the same site as clone 29c, but instead had an unusually long 3'-noncoding region containing three possible polyadenylation signals with no poly(A) tail. The differences in the 3'-noncoding region could represent alternate splicing, allelic variants, or possibly separate genes. However, these clones are designated as allelic variants of (IIC18) because they differ by only one base in the coding region. They are most similar to 254c (IIC17) (86% amino acid homology) and less similar to IIC9 (82%) and IIC8 (77%) (Table I). A third unique P450IIC cDNA, clone l l a (designated IIC19), was also identified. IIC19 is most homologous to 254c (IIC17) (95% nucleic acid homology), but base differences were found throughout both the coding and noncoding sequence (Figure 2 and Table I). It is 92% homologous in its amino acid sequence to both IIC17 and IIC9,81% homologous to IIC18, and 79% homologous to IIC8. Clone 1l a had a short 5'-leader sequence and contained the stop codon, but did not have a polyadenylation signal or poly(A) tail. Interestingly, no clones for IIClO (MP-8) were isolated from either library, despite the sequencing of the 3' region of all 50 putative IIC9 clones. Figure 4 shows the alignment comparisons for the deduced amino acid sequences of all known members of the human CYPZC family, including the three new P450s identified in the present study. The 7 proteins, along with the consensus sequence, can be aligned with no gaps, and each is predicted to be 490 amino acids long. The amino acid sequences show marked similarities with many regions of absolute conservation. 'Regions of marked conservation are noted from 131 to 180, and from 302 to 460. These human P450IIC protein sequences also demonstrate hypervariable regions which may be important for interactions between the enzyme and substrate. These include the region from 181-210 and 220-248 as well as 283-296 and a short region near the carboxy terminus at 461-479. Notably, a putative recognition site for phosphorylation of P450 by CAMP-dependent kinase for P4502B1 (Muller et al., 1985) (Arg-Arg-Phe-Ser) at positions 124-127 was conserved in IIC8, IIC9, and 1 la (IICl9), suggesting that these cytochromes might be regulated by phosphorylation. However, neither IIC17 nor IIC18 contained a serine at this site. The overall percent homology for both nucleic acid and protein sequences is summarized in Table I.

Cytochrome P450IIC Subfamily Table 111: Mephenytoin Hydroxylase Activity by P45011C cDNAs Expressed in Microsomal Fractions of COS-I Cells' racemic mephenytoin 4-hydroxylase act. [pmol cDNA transfected m i d (mg of protein)-'] controls 14 i 5 65 (IIC9) 12 25 (11C9) 19 16 Ila (llC19) 29c (IIC18) 86 & 22b 6b (llC18) 46 i 6b Cells were transfected with the cDNA indicated. Incubations were performed with 200 pg of microsomal protein, 0.5 mM substrate, 1000 units of reductase, 0.2 nmol of cytochrome b5,and 2 mM NADPH in a final volume of 0.5 mL. All values are the mean f SE of 6-9 determinations except for 65, 25, and 1 la which are the average of duplicate values. bSignificantly different from controls, P < 0.01 by the MannWhitney U test. These values compare with values of 239 i 51 pmol min-' (mg of protein)-' for human liver microsomes.

cDNA Expression in COS-1 Cells. The two variant cDNAs for TIC9, the two variant cDNAs for IIC18, and the cDNA for IICl9 were inserted into expression vectors and transfected into COS-I cells. Cell lysates were prepared and immunoblotted by using antibody to HLx and P4501IC9. The results are shown in Figure 5 . Transfection of COS- 1 cells with the two variants of 1TC9 (25 and 65) resulted in the expression of a protein with a molecular weight equal to that of pure IIC9. In contrast, neither TIC18 (either variant) nor IIC19 was detected by antibody to HLx or IIC9. However, Northern blot analysis indicated that all three cDNAs had been successfully transfected into these cells (Figure 6). The sizes of the transcripts were those expected for the constructs. The somewhat lesser hybridization of the IIC oligoprobe with RNA from cells transfected with l l a reflects a lower amount of RNA in this sample as shown by the hybridization with the actin probe. Table 111 compares the metabolism of racemic mephenytoin in COS-I cells transfected with the two allelic variants of IIC9, two variants of ITC18, and IIC19. Only COS-1 cells transfected with cDNAs for IIC18 (both allelic variants) exhibited a significant increase in the metabolism of mephenytoin over that of nontransfected cells. Values for IIC9 metabolism represent duplicate samples of both clones. However, when the four values for both clones are pooled, they are not significantly different than metabolism in untransfected COS-1 cells. Human Tissues. Expression of P450IIC8 was examined in mRNA from various human hepatic tissues using Northern blot analysis with the specific oligoprobe for IIC8 (T300R). Hybridization with T300R was negligible in mRNA from 860624 compared to S33 and a number of other liver samples (Figure 7). When corrected for hybridization with the actin probe, the amounts of IIC8 mRNA were consistent with the relative amounts of HLx observed in Western blot analysis. Laser scans of the radioautographs indicated that IIC8 mRNA levels in sample 860624 were at least 70-fold lower than S33 and 3-1 5-fold lower than any of the remaining samples. DISCUSSION Genetic analysis of the CYPZC subfamily indicates the presence of 2 7 genes. The present study characterizes the profile of cDNAs from this subfamily in a library from a single individual. The majority of the cDNAs characterized in the high-HLx library (60%) were one of two allelic variants of IIC9, while 35% represented IIC8. Two new genes were identified (two allelic variants of IIC18 and IIC19), and a third new gene (IIC17) was identified in a library from a second

Biochemistry, Vol. 30, No. 13, 1991

3251

individual. A comparison of the P450IIC cDNAs and their predicted amino acid sequences shows that 70% of the amino acids are absolutely conserved among the human P450IIC subfamily. Regions of the human P450IIC protein sequence have particularly enhanced conservation, and these areas may participate in common P450 functions. The hypervariable regions may be responsible for substrate specificity. However, changes of only one amino acid have been shown to alter the substrate specificity of P450coh from coumarin hydroxylase to P45015 a-steroid hydroxylase activity (Lindberg & Negishi, 1989). Therefore, small changes in amino acid composition can completely reverse substrate specificity. A putative recognition site for phosphorylation of P450 by CAMP-dependent kinase (Muller et al., 1985) was found in IIC8, IIC9, and IIC19, suggesting that catalytic activity of these cytochromes might also be regulated by phosphorylation. In contrast, neither IIC17 nor IICl8 contained the phosphorylatable serine at position 127. There are a number of polymorphisms in the human CYPZC subfamily. These include variations in the hepatic levels of HLx (Wrighton et al., 1987) and metabolic variations in the hepatic metabolism of (S)-mephenytoin. The molecular basis for these polymorphisms has not been characterized. IIC8 appears to encode the protein for HLx on the basis of its N-terminal amino acid sequence (Okino et al., 1987; Wrighton et al., 1987; Lasker et al., 1987). In the present study, 29 IIC8 clones were identified (-40 000 clones screened) in an individual who was phenotypically high for hepatic HLx expression. In contrast, no IIC8 clones were identified in the library from the individual who expressed extremely low amounts of HLx (>90 000 clones screened). These results were consistent with those of Northern analysis which indicate that IIC8 mRNA was low or absent in this individual. This suggests that the low amounts of HLx may result from a gene defect which results in defective expression of the mRNA in the low phenotype and not from a defective mRNA. At least two P450 proteins which metabolize (S)-mephenytoin have been isolated from human liver (MP-1 and MP-2). cDNAs for IIClO and IIC9 (which vary by only two amino acids in the coding region) were originally isolated by using antibody to P45oMp.l. However, neither the hepatic content of P45OMP.,or P45oMp.z nor the expression of mRNAs for MP-8 (IIClO), MP-4 (IIC9), or IIC8 could be correlated with mephenytoin hydroxylase activity (Ged et al., 1988). A number of allelic variants of IIC9 have been characterized by several different laboratories. We have isolated two additional full-length allelic variants of IIC9. One of these clones is identical with MP-4, but is full-length. It varies from the almost full-length human form 2 isolated by Yasumori et al. (1987) by only two silent base changes in the coding region and by four changes in the noncoding region. The number of differences in the nucleic acid sequences of the presumed allelic variants isolated by different laboratories range from 4 to 17 and the amino acid changes vary from 0 to 4, as illustrated in Figure 3. Two of the amino acid differences occur within the first six N-terminal residues, the others occurring singly throughout the sequence. The effect of these changes on catalytic activity has not been systematically studied. Thus far, Relling et al. (1990) found that when the cDNAs for IIC8 and IIC9 isolated by Kimura et al. (1987) were expressed in HepG2 and TK- cells, both cytochromes metabolized tolbutamide (IIC9 > IIC8) and IIC9 4hydroxylated racemic mephenytoin but did not metabolize (S)-mephenytoin. However, Yasumori, et al. (1990) found that the form of IIC9 isolated in his laboratory (human form

3252 Biochemistry, Vol. 30, No. 13, I991 29c

65

Romkes et al.

-198

-100

GGCACCGGAA AGAACAAGAA AAAAGAACAC CTTATTTTTA TCTTCTTCAG TGAGCCAATG TTCATTCAAA AGAGAGATTA AAGTGCTTTT TGCTGACTA -99

-1

G AAGGCTTCA GTCACAGTCA GAGTCAGAAT CACAGGTGGA TTAGTAGGGA GTGTTATAAA GCCTTGAAGT GAAAGCCCGC AGTTGTCTTA CTAAGAAGAG AAGCCTTCA 6b AGT GAAAGCCCGC AGTTGTCTTA CTAAGAAGAG AAGCCTTCA lla CTTCA

29c

_.

QQ

65 ATGGATTCTC TTGTGGTCCT TGTGCTCTZT CTCT”ATG?I TGCTTCTCCT TTCACTCTGG AGACA‘AXT AT‘XITACCAG CTGT‘XCTCT GGT‘XTCTZT CTCTCCTGTT TGTTCTCCT TT”ACTCTX A‘rGCAGAGCT 6b ATXTACCAG CTGTGGCTCT GGTKTCTGT CTCTCCTGTT TGTTTCTCCT TTCACTC:GG AGtCA‘ALXT lla ATGGATCCTT TZTGGTCCT TGTL-TCTGT CTCTCATSTT 7-XTTCTCCT TTCAATCTGG AGA”A”AG-T

29c

CTGGGA‘AGG AAAACTCCCT CTGGAAGGG ”AGGCTCCCG CTGGARGAGG ZA‘XTCCCG CT*-*AX AAAACTCCCT

CCT-XCCCC

TCT-XCCCC TCT“3XCCC CCTGGiCCf

100 198 65 ACTCCTCTCC CAGTGATTGG AAATATCCTA CAGATAGGTA TTAAGGACAT CAGCAAATCC TTAACCAATC TCTCAAAGGT CTATGGCCCT GTGTTCACT 2 9c

ACTCCTCTCC CGATTATTGG AAATATCCTG CAGTTAGATG TTAAGGACAT GAGCAAATCC TTAACCAATT TCTCAUAGT CTATGGCCCT GTGTTCACT 6b ACTCCTCTCC CGATTATTGG AAATATCCTG CAGTTAGATG TTAAGGACAT GAGCAAATCC TTAACCAATT TCTCAAAAGT CTATGGCCCT GTGTTCACT lla ACTCCTCTCC CAGTGATTGG AAATATCCTA CAGATAGATA TTAAGGATGT CAGCAAATCC TTAACCAATC TCTCAMAAT CTATGGCCCT GTGTTCACT 65

2 9c 6b lla

-__

298

79C

65 TTCCCACTGG CTGAAAGAGC TAACAGAGGA TTTGGAATTG TTTCCAGTGG CTWWIRAGT TAACAAAGGA CTTGGAATCC 6b TTTCCAGTGG CTWVIAAAGT TAACAAAGGA CTTGGAATCC lla TTCCCACTGG CTGAAAGAGC TAACAGAW TTTGGAATCG

TTTTCAGCAA TGGAAAGAAA TGGAAGGAGA TCCGGCGTTT CTCCCTCATG ACGCTGCGG

29c

TTTTCAGCAA TGGARRGAGA TGGAAGGAGA TCCGGCGTTT CTGCCTCATG ACTCTGCGG

254c

TTTTCAGCAA TGGAAAGAGA TGGAAGGAGA TCCGGCGTTT CTGCCTCATG ACTCTGCGG TTTTCAGCAA TGGAAAGAGA TGAAGGAGA TCCGGCGTTT CTCCCTCATG ACGCTGCGG GGAAGGAGA TCCGGCGTTT CTGCCTCATA ACTCTGCGG

391

495

65 AATTTTGGGA TGGGGMGAG GAGCATTGAG GACCGTGTTC AAGAGGARGC CCGCTGCCTT GTGGAGGAGT TGAGARAARC CAAGGCCTCA CCCTGTGAT

TGGGGAAGAG GAGCATCGAG TGGGGAAGAG GAGCATCGAG TGGGGAAGAG GAGCATTGAG TGGGGAAGAG GAGCATCGAG

GACCGTGTTC GACCGTGTTC GACCGTGTTC GACCGTGTTC

AAGAGWIRGC AAGAGGARGC AAGAGGAAGC AAGAGGAAGC

CCGCTGCCTT CCGCTGCCTT CCGCTGCCTT CCGCTGCCTT

GTGGAGGAGT GTGGAGGAGT GTGGAGGAGT GTGGAGGAGT

TGAWIAARAC TGAGARAARC TGAWIAARAC TGAGARAARC

ULATGCCTCA CAATGCCTCA CAAGGCTTCA CAATGCCTCA

CCCTGTGAT CCCTGTGAT CCCTGTGAT CCCTGTGAT

65 GAAAAGTTGA ATGAAAACAT CAAGATTTTG GAAAAATTCA ATGAAAACCT CAGGATTCTG 6b WWIRATTCA ATGAAAACCT CAGGATTCTG lla GAAAAATTGA ATGAAAACAT CAGGATTGTA 254c WWIRATTGA ATGAAAACA? CAGGATTGTA

AGCAGCCCCT AGCTCTCCAT AGCTCTCCAT AGCACCCCCT AGCACCCCCT

GGATCCA‘rAT GGATCCAGGT GGATCCAGGT GGATCCAGAT GGATCCAGAT

CTGCAATAAT CTGCAATAAT CTGCAATAAT ATGCAATAAT ATGCAATAAT

TTTTCTCCTA TCATTGATTA TTCCCTGCTC TCATCGATTA TTCCCTGCTC TCATCGATTA TTTCCCACTA TCATTGATTA TTTTCCACTA TCATTGATTA

CTTCCCGGGA TCTCCCAGGA TCTCCCAGGA TTTCCCGGGA TTTCCCGGGA

ACTCACAAC AGTCATAAT AGTCATAAT ACCCATAAC AGTCATAAC

TATGCTCTCC TATGGACTCC TATGGACTCC TATGCTCTCC TATGCTCTCC

TTCTCCTGCT GAAGCACCCA TGCTCCTGCT GAAGTACCCA TGCTCCTGCT GAAGTACCCA TTCTCCTGCT GAAGCACCCA TTCTCCTGCT GAAGCACCCA

GAGGTCACAG CTAAAGTCCA GAGGTCACAG CTAAAGTCCA GAGGTCACAG CTAAAGTCCA GAGGTCACAG CTAAAGTCCA GAGGTCACAG CTAAAGTCCA

GWLAGAGATT GAACGTGTG GGAAGAGATT GAATGTGTA GGMGAGATT GAATGTGTA GGARGAGATT GAACGTGTC GGMGAGATT GAACGTGTA

2 9c

AATTTTGGGA 6b AATTTTGGGA lla AATTTTGGGA 254c AATTTTGGGA

65 29c

6b lla 254c 595 29c

693

65

2 9c

6b lla 254c 25

65 2 9c

6b lla 254c 892

GGGACAGAGA CGACAAGCAC AACCCTGAGA GGAACAGAGA CAACGAGCAC CACTCTGAGA 6b GGAACAGAGA CAACGAGCAC CACTCTGAGA lla GGWLCAGAGA GCACAAGCAC AACCCTGAGA 254c GGAACAGAGA CAACAAGCAC AACCCTGAGA 65 29c

990

43: 1C83 65 ATTGGCAW ACCXAGCCC CTLXRTGCAR GACAXAGCC AZATXCCTA CACAZATXT STXTGCACG AGGTCCAGAG ATAGTGAC CTTCTCCCC

25

2TT‘X;CAGAA ACCGjAG-CC CTGTATGCAG ACC‘XAGCCC CTZTATGCAG 6b GTT‘XCA‘9A lla ATTGGCA‘A ACC”XAGCC’3 CTSCATGCAG 254c STTLm-2A ACCGGA”XCC CT”XAT‘;CAZ 2 9c

-ACA‘XAGTC A-ATZCCCTA ^ACA”AT‘XT GACAGGASTC ACATGCCCTA CACAGATGCT GA-A-GGCC ACATXCCTA -ACA-ATLXT GACAGGA‘XC ACATGCCCTR CACAGAT‘XT

ZT‘XIT-XACS AGATC‘AGA.3 GTGGTGCACG AGATCCAGAS ST‘j;TXhCS A‘XiTCCA‘AS ZTZZTGCACS A”XiTX4-AG

CTCCTCCCC ATA”ATTGAC CTCCTCCCC ATACATC-AC CTCATCCCC ATR‘ATTGAC CTCATCCCC ATACATTGAC

:c5:

1186

65 ACCAGCCTX CCCATGCAST ‘3CCTGT”AC ATTRRRTT-A ‘IRRRCTATC: ‘ATTCCCAAG -&-ACAACCA

TATTAATTC CCTGACTTCT GT-XTA-AT ‘”A TAATAACATC CCTGACTTCT ZTLXT‘XAC ACCAACCTGC CCCATGCAZT GACCTGTGAT GTTRRRTTCA RAAACTACCT “ATCCC‘IAAG 6b ACZAACCTGC CCCATXAGT SACCTGTGAT GTTPAATTCA AAAACTACCT CATCCCZAAG S E C A TAA-AACATC CCTGACTTCT GTXTXAC 1la ACCA‘KCTGC CCCAT-XAGT GACCTSTGAC GTTAAATCA GAAACTACCT CATTCC“4AZ “XCAr3ACCA TATTAACTC CCTCACTTCT GTXTACAT ?54C AC”ArXCTX CCCAT‘XAGT ‘ACCTGTGAC Sl?AAATTrA ‘IRRRCTACCT CATTCC-IAAZ GGCA”IAACCA TATTAACTTC CCTCACTTCT GT”XTA”AT 29c

CCCAGAGATG CCCAGAGATG CCCAGAGATG CCCAGAGATG CCCAGAGATG

TTTGACCCTC TTTGACCCTG TTTGACCCTG TTTGACCCTC TTTGACCCTC

ATCACTTTCT GCCACTTTCT GCCACTTTCT GTCACTTTCT GTCACTTTCT

1287 GGATGAAGGT GGCAATTTTA AGAAAAGTAA ATACTTCATG ...~. .. .. .... . .. CrTTTrTCA . -. .. . .. GGATAAGAGT GGCAACTTTA AGAAAAGTGA CTACTTCATG CCTTTCTCA GGATAAGAGT GGCAACTTTA AGAAAAGTGA CTACTTCATG CCTTTCTCA GGATGAAGGT GACAATTTTA AGAAAAGTAA CTACTTCATG CCTTTCTCA GGATGAAGGT GGAAATTTTA AGAAAAGTAA CTACTTCATG CCTT-A

GGATTTGTGT GGGAGAAGCC GGATGTGTAT ‘XGAGAGGGC GGATGTGTAT GGGAGAGGGC GGATTTGTGT GGGAGAGGGC

C‘.’GGCCGGCA CTGGCCCGCA CTGGCCCGCA CTGGCCCGCA

TGGAGCTGTT T-LXAGCTGTT TGGAGCTGTT TGGA-XTGTT

TTTATTCCTG TTTATTCCTG TTTATTCCTG TTTATTCCTG

1189

65

GACAACAAAG AATGACAAAG 6b AATGACAAAG lla GACAACAAAG 2 5 4 c GACAACAAAG 29c

65

29c

6b lla

1288 GCAGGAAAAC GCAGGARRAC “XAGGAUAC ”XAGGAAAAC

AATTTCCCAA AATTCCCCAA AATTCCCCAA AATTTCCCAA AATTTCCCAA

~

ACCTCCATTT ACCACCATT ACCACCATTT ACCTTCATTT

~~~

TACAGAACTT TGCAGARCTT TGCA-ACT? TACAGAACTT

1386 ~... TAACCTGAAA TCTCTGGTT TAACCTGAAA TCTCAGGTT TAACCTGAAA TCT-JGGTT TAACCTr& TCTCTGATT

254c 25 65 29c

6b lla 254c

GACCCAAAGG ATATTGACAT GACCCAAAGG ATATTGACAT GACCCAAAGG ACCTTZACAC r uA-CCAAAGG ACCTTGACAC

CACCCCATTG CCAATGCAFT CACCCCATTG CCAATGCATT AACTCCTGTT ZTCAATGGAT AACTCCTGTT GTCAATGGAT

TGGTCGGTGT GCCACCCTTG TGGTCGGTGT GCCACCCTTG TXCTTCTGT CCCGCCCTTC TTGCTTCTST CCCGCCCTTC

TACCAGCTCT GCTTCATTCC TGTCTTdCA AC&GCA.OAT TACCRGC+CT GciTi~iTiCi i T C + ~ S Z AA G r‘ GCAGT TATCAGCTST GCTTCATTCC TGTCXSAGA AGCACA-AT TATCAGCTGT XTTCATTCC TSTCT-AASA A-LA-JGAT

GTCCCTGCAG GTCACCTGCA GTCACCTGCA GTCCCTGCAG GTCCCTGCAG

TCTGGGGCAT TATCCATCTT ATCAGGGCCA TTAGCCTCTC ATCAGGGCCA TTGGCTCTCC TCTGGTCCAA ATTTCACTAT TCTGGTCCAA ATTTCACTAT

TCACTATCTG CCTTCTCTCT CTTCTCTCTA CTGTGATGCT CTGTGATGCT

TAATGCCTTT TCTCACCTGT GTGAGGGATA TTTTCTCTGA TGAGGGATAT TTTCTCTGAC TCTTCTGACC CGTCATCTCA TCTTCTGACC CGTCATCTCA

TTACGGAGAG CCTCCATTAA CCTCCATTAA TTCACTGTGC AGTTTCACGG

TTCACTGTGC TTGGGTCACT TTGGGTCACT GCTATTCCCA

1683 AAATATATCT GCTATTCTCC ATACTCTGT TCCTAAATAT ATCTGCTATT CTCCATACT TCCTAAATAT ATCTGCTATA ACTCCATAC TACTCTATAA TAGTTACATT GAGTGCCAC

1486

65

GGCCTGGCTG AGTTTGGCTG 6b AGTTTGGCTG lla GGTCTGGCTG 2 5 4 c GGTCTGGCTG 29c

1584

CTGCTGTGCA CTCCTGTGCT CTCCTGTGCT CTCCTGTGCT CTCCTGTGCT

CTCTCTTTCC ATTCTCCCTT ATTCTCCCTT CTCTCTTTCC CTCTCTTTCC

1585 65 29c

6b lla 254c

TTTTCCCTTC CCTGAAGATC TAGTGAACAT TCGACCTCCA CCACATCTTC CCATTCCCTC AAGATCCAAT GAACATCCAA CACATCTTCC CATTGCCTCA AGATCCAATG AACATCCCAA TTCCCCCAAG ATCTAGGAAC ATTCAGCCTC CATTAAAAAG TTCCCCCAAG ATCTAGTGAA CATTCAGCCT CCATTAAAAA

TTTCCTATGT AGAGAGTTTC AGAGAGTTTC AAATATATCT

CATCTCACA CTTGTCAAT TTGTCAATC CATTTTCCC CATTTTCCC

Biochemistry, Vol. 30, No. 13, 1991 3253

Cytochrome P450IIC Subfamily

1780 TATTAATATG TTATTATTAA ATAGAGAAAT ATGATTTGTG TATTATAAT AGTTGTCAGT ATGTTATCAC TAGAAAACAA AGAAAAATGA T T W T GAGTTGTAGA GTTATTACTA TAAACAAAGA AAAATGATT-GCG TAGA

65 29c 6b lla

1684 AACAGTTGCA CTGTATCACT TCTGTATCAC ATAATGCTGA

65 29c 6b

1781 1881 TCAARGGCAT TTCTTTTCTG CATGTTCTAA ATAAARAGCA TTATTATTTG CTGAAAAAAA GACAATTCAG AGCCARAAAA AWAAAAA T T A G G C A G AAAGGTCAAG AGGTAGARAG CTGTTTAGCT AAATGCCACC TAGAGTTATT GGAGGTCTGA ATTTGAAAAA AAAACTATGT CCAGGAGCA

TTGACTGTCA TGTATTGACC TTGATTGACC TACTTGTCTA

CATAATGCTC ACCACATATG ACCACAGATT ATGTTGAGTT

ATACTTATCT CTAATACCTA GCTAATACCT ATTAACATAT

AATGTTGAGT TCTACTGCTG ATCTACTGCT TATTATTAAA

1882 1980 6b GCTGTAACCT GTAGCWLAAT AATGGAAWLA TCATCCATAA GAGGGATGAA CATTAAGTGT TTGAATTCAT GCTCTGCTTT TGTGTTACTG TAAACACAA

2053 TGCTTTCATA TGG

1981

6b GATCAAGATT TGWLTAATCT TTTTCCTTTG TGTTTCCAAC TTAGATCATG TCTAAA-

FIGURE 2: Nucleotide sequences of human P450IIC cDNAs. Sequences were determined by the dideoxy chain termination method. The differences observed for clone 25 are listed above the sequence for 65 and are boxed. The initiation and termination codons are starred. The heme binding region and polyadenylation signals are underlined. The one-base difference between 29c and 6b is also boxed. CONSEN IICl Hum2 Meehan

1 100 PIDSlVvLVLC L S C L L L L S L W R Q S S G R G K L P P G P T P L P V I G N I L Q I G I K D I S K S L T N L S K V Y G P V F T L Y F G L K P I V V L H G Y E A V K E A L I D L G E E F S G R G I F I S XXX

65

25 np-8 CONSEN IICl Hum2 HP-4 Heehan 65 25 MP-8 CONSEN IICl Hum2 np-4 Meehan 65 23 MP-8 CONSEN IICl Hum2 HP-4 neehan 65 25 flP-8 CONSEN IICl Hum2 np-4 neehan 65 25 np-8

xxxxx 101 200 PLAERANRGF GIVFSNGKKW KEIRRFSLHT LRNFGMGKRS IEDrVQEEAR CLVEELRKTK ASPCDPTFIL GCAPcNVICS IIFHKRFDYK DQQFLNLMEK C

xxxxxxx C

Y

201 300 LNENIKILSS PWIQICNNFS PIIDYFPGTH NKLLKNVAfH KSYILEKVKE HQESHDMNNP QDFIDCFLMK MEKEKHNQPS EFTIESLENT AVDLFGAGTE

L

301 400 TTSTTLRYAL LLLLKHPEVT AKVQEEIERV IGRNRSPCMQ DRSHMPYTDA VVHEVQRyiD LLPTSLPHAV TCDIKFRNYL IPKGTTILIS LTSVLHDNKE

L C 401 490 FPNPEWFDPH HFLDEGgNFK KSKYFMPFSA GKRICVGEAL AGMELFLFLT STLQNFNLKS LVDPKNLDTT PVVNGFASVP PFYQLCFIPV

'

D

Alignment of human IIC9 proteins. The deduced amino acid sequences for hum2 (human form 2; Yasumori et al., 1987), MP-8, and MP-4 (Ged et al., 1988) are not complete, and the alignment begins following the 'X". The other previously published IIC9 sequences are IICl (Kimura et al., 1987) and PB-1 (Meehan et al., 1988). Invariant residues are indicated in the consensus sequence as capitals, while residues which differ are indicated below and are also listed in lower case letters in the consensus sequence. The termination codon is starred. Therefore, we would not expect to detect an increase in me2) metabolized (S)-mephenytoin preferentially when expressed phenytoin metabolism in IIC9-transfected COS- 1 cells in yeast. These forms differed by only three amino acids. In equivalent to the low turnover numbers reported by Yasumori contrast, Brian et al. (1989) found that when a full-length et al. (1990) for IIC9 in yeast [14 pmol min-' (nmol of MP-8 (constructed with the first 15 nucleotides predicted from P450)-']* the known amino acid sequence of P45OMP.,)was expressed However, two allelic variants of IICl8 did 4-hydroxylate in yeast, it did not metabolize (S)-mephenytoin. This form racemic mephenytoin when inserted into COS-1 cells in the would differ from human form 2 by only two amino acids. Thus, the role of TIC9 in (8-mephenytoin metabolism remains present study at a rate significantly higher than controls, suggesting that these cytochromes may be important in mecontroversial. phenytoin metabolism. Although we could not measure the In the present study, expression of the two variant full-length concentration of IICl8 in COS-I cells in the absence of an clones (65 and 25) of IIC9 in COS-1 cells resulted in exantibody for this cytochrome, if we assume that the concenpression of a protein that reacts with antibody to a purified tration is similar to that of IIC9 in COS-1 cells transfected form of P450 which metabolizes mephenytoin in a reconstiwith that cDNA (1 3 pmol/mg of protein), the rate of racemic tuted system (anti-IIC9). However, neither of these transmephenytoin metabolism by IIC 18 would be approximately fected cells appeared to metabolize racemic mephenytoin at 4000 pmol min-' (nmol of P450)-'. This estimated rate is a rate appreciably above that of nontransfected COS-1 cells -100 times higher than the rate of (S)-mephenytoin me[14 pmol m i d mg of protein)-']. Yasumori et al. (1990) tabolism reported for IIC9 (Yasumori et al., 1990) when reported that the cDNA for human form 2 (differing from expressed in yeast. Additional studies in a cDNA expression clone 65 by only two silent base changes in the coding region) system which results in higher levels of expression of P450 metabolized (S)-mephenytoin when expressed in yeast. protein will be required to resolve the exact relationship of the However, expression of the P450 IIC9 protein in COS-1 cells various allelic variants of IIC9, IIC18, and other CYPZC was estimated to be 13 pmol/mg of microsomal protein, while the expression of IIC9 in yeast was 10 times higher. proteins to the metabolism of mephenytoin (particularly the FIGURE 3:

-

-

3254 Biochemistry, VoI. 30, No. 13, 1991

Romkes et al.

1 )Idp.-

IIC8 65 25 29c 6b lla

IIC8 65 25 29c 6b lla

100

CScllLlSlU msgRgkLP -.I6 UiIiQid.KDi sKSll".SKv H E P R N L W LSFMLLF'SLM RQSCRRRKLP PGPTPLPIIG -1DVKDI CKSFTNFSKV MDSLVVLVIL~SCLLLLSLW RQSSGRGKLP PGPTPLPVIG NIIQIGIKDI SKSLTNLSKV I D S L W L W LSCLLLLSLW RQsSGRGKLP PGPTPLPVIG NIIQIGIKDI SKSLTNLSKV MVPAVALVIX: LSCLFLLSLW RQSSGRGRLP SGPTPLPIIG NILQLDVKDM SKSLTNFSKV W V P A V A L W LSCLFLLSLW RQSSGRGRLP SGPTPLPIIG NILQLDVKDM SKSLTNFSKV MDPF7NLVLC LSCLLLLSIW ROSSGRGKLP PGPTPLPVIG NILQIDIKDV SKSLTNLSKI 101 P1aeran.G. PISQRITKGL PLAERANRGF PLAERANRGF PVAEKVNKGL PVAEKVNKGL PLAERANRGF

254c

-.YE

IkpiWlRGY YGPVFTVYFG MNPIWFHGY YGPVFTLYFG LKPIWLHGY YGPVFTLYFG LKPIWLHGY YGPVFTVYFG LKPIWLHGY YGPVFTVfFG LKPIWLHGY YGPWTLYFG LERMWLHGY

v r CLVEBLRKTk m L OCAPCMCS (nvfSWKrU-msLmt IEDRVQEEAH CLVEEw(TK ASPCDPTFIL GCAPCNVICS GIISSNGKRW K E I W S L T N SGIVFSNGKKW REIRRFSWT L R N F m S IEDRVQEEAR CLVEELRKTK ASPCDPTFIL GCAPCNVICS GTVFSNGKKW lCEIRRPSWT LR"cRs IEDRVQEEAR CLVEELRKTK ASPCDPTFIL GCAPCNVICS GILFSNGKRW KEIRRFTWT LRNFQ4GKRS IEDRVQEEAR CLVEELRKTN ASPCDPTFIL GCAPCNVICS IEDRVQEEAR CLVEEW(TN ASPCDPTFIL GCAPCNVICS G1LFSbX;ltRW KEIRFIFCWT GIVFSNGKRW I[EIRRFSWT LRNFQ4XR.S IEDRVQEEAR CLVEELRKTK ASPCDPTFIL GCAPCNVICS KEIRRFCLIT LRNFGMGKRS IEDRVOEEAR CLVEELRKTN ASPCDPTFIL GCAPCNVICS

EaVlCLUtIDl m.f EAVKEALIDN GEEPSGRGNS

EAVICEALIDL GEEFSGRGIF EAVllWIDL EAVICEALIDH EAVKEALIDH EWREALIDL

GEEP-IF GEEFSGRGSF GEEFSGRGSF GEEFSGRGHF

200 iiFhkRFDYK DQqRLnMek VVFQKRFDYK DQNFLTIMKR IIFHKRFDYK DQQFLNLMEK IIFHKRFDYK DQQFLNIMEK VIFHDRFDYK DQRFLNLMEK VIFHDRFDYK D Q R F L N W K IIFQKRFDYK DQQFLNLMEK I iFHKRFDYK DOQFLNLMEK 300

201

254c

UBLllirllss FNENFRILNS LNENIKILSS LNENIKILSS EWENLRILSS E'NENLRILSS LNENIRIVST LNENIRIVST

IIC8 65 25 29c 6b

301 t"wmerV TTSTTLRYGL LLLtKHPEVT "STTLRYAL LULKHPEVT TTSTTLRYAL LULKHPEVT rrSTTLRYGL LLLLKYPEWT TTSTTLRYGL LLLLKYPEVT

IIC8 65 25 29c 6b lla 254c

401 m e - g FPNPNIE'DPG FPNPEME'DPH FPNPEMFDPH -EMX?DPG FPNPEKF'DPG FPUPEMFDPR FPNPEKEQPR

IIC8 65 25 29c 6b lla

mic196p PWIQVCNNFP PWIQICNNFS PWIQICNNE'S PWIQVCNNE'P PWIQVCNNE'P

.iIDyfPGtH lOltllklhrafm kSyilEkvl[d HQeSlDmNnp rDFiDCFLiK WqEKhNQqS EFtiEsLiaT vtDlfgAGTG

LLIDCFPGTH PIIDYFPGTH PIIDYFPGTH ALIDYLPGSH ALIDYLPGSH PWIQICNNE'F TIIDYFPGTH PWIQICNNFS TIIDYFPGSH

NKVLKENALT NKLLKNVAFM NKLLKNVAFM NKIAENFAYI NKIAENFAYI NKLLKNLAFM NKLLKNLLTL

RSYIREKVKE KSYILEKVKE KSYILEKVKE KSYVLERIKE KSYVLERIKE ESDILEKVKE KSYVLERIKE

HQASLDVNNP HQESMDMNNP HQESMDWP HQESLDMNSA HQESLDMNSA HQESMDINNP HQESLDMNNP

bRsDI" IGRHRSPCMQ DRSHMPYTDA WHEIQRYSD IGRNRSPCMQ DRSHMPYTDA VVHEVQRYID IGRNRSPCMQ DRSHMPYTDA WHEVQRYLD VGRNRSPCMQ DRSHMPYTDA WHEIQRYID VGRNRSPCMQ DRSHMPYTDA WHEIQRYID IGRNRSPCMQ DRGHMPYTDA WHEVQRYID VGRNRSPCMQ DRSHMPYTDA WHEVQRYID

-1

AKVQEEIDHV AKVQEEIERV AKVQEEIERV AKVQEEIECV AKVQEEIECV l l a STSTTLRYAL LLLLKHPEVT AXVQEEIERV 2 5 4 ~ T T S ~ Y A LLLLLKHPEVT AKVQEEIERV

~ e g g ~ ~ c B F M W - iSC *Rg f Y + R r m

HFLDKNGNE'K KSDYFMPFSA HFLDEGGNFK KSKYE'MPFSA HFLDEGGNE'K K S K Y W F S A HFLDKSGN"K KSDYFMPFSA HFLDKSGNFK KSDYFHPFSA HFLDEGDNFK KSNYFMPFSA HFLDEGGNFK KSNYFMPFSA

EFNIENLVGT EFTIESLENT EFTIESLENT EFTVESLIAT EFTVESLIAT EFTIESLVIT EFTVESLIAT

VADLFVAGTE AVDLFGAGTE AVDLFGAGTE VTDMFGAGTE VTDMFGAGTE AADLLGAGTE VTDMFGAGTE

400 LlPTslPHAV TcDvKFrNYL IPKGtTIlts L T W d n K E LVPTGVPHAV TTDTKFRNYL IPKGTTIMAL LTSVLHDDKE LLPTSLPHAV TCDIWRNYL IPKGTTILIS LTSVLHDNKE LLPTSLPHAV TCDIWRNYL IPKGTTILIS LTSVLHDNKE LLPTNLPHAV TCDVKFKNYL IPK IITS LTSVLHNDKE LLPTNLPHAV TCDVKFKNYL IPK%ITS LTSVLHNDKE LIPTSLPHAV TCDVWRNYL IPKGTTILTS LTSVLHDNKE LIPTSLPHAV TCDVKFRNYL IPKGTTILTS LTSVLHDNKE

1vDpKdldtT TILQNFNLKS VDDLKNLNTT SILQNFNLKS LVDPKNLDTT SILQNFNLKS LVDPKNLDTT TILQNFNLKS QVDPKDIDIT TILQNE'NLKS QVDPKDIDIT FILQNFNLKS LIDPKDLDTT FILQNF'NLKS LIDPKDLDTT

-t

GKRICAGEGL ARMELFLFLT GKRICVGJDL AGMELFLFLT GKRICVGEAL AGMELFLFLT GKRMCMGEGL ARMELFLFLT GKRMCMGEGL ARMELF'LFLT GKRICVGEGL ARMELFLFLT GKRICVGEGL ARMELFLFLT

RDFMDCFLIK MEQEKDNQKS QDFIDCFLMK MEKEKHNQPS QDFIDCFLMK MEKEKHNQPS RDFIDCFLIK MEQEKHNQQS RDFIDCFLIK MEQEKHNQQS QDFIDCFLIK MEKEKQNQQS RDFIDCFLIK MEQEKHNQQS

pvvngfasvP AVTKGIVSLP PWNGFASVP PWNGFASVP PLPMHLVGVP PLPMHLVGVP PVVNGFASVP PVVNGFASVP

491 PfYQlCFIW * PSYQICFIPV * PFYQLCFIPV * PFYQLCFIPV * PLYQLCFIPV * PLYQLCFIPV * PFYQLCFIPV * PFYQLCFIPV *

FIGURE 4: Alignment of human P450IIC proteins. The consensus sequence is indicated in the top line, where conserved residues are capitalized and shaded in gray and variable residues are indicated by dots or listed as lower case. The heme binding region is underlined. The termination codon is starred.

anti-IIC9

IIC500R

Actin

FIGURE 5:

Western blot of recombinant transformed COS-1 cells. Each lane represents microsomal protein (50 pg) from an independent transformation with the indicated P45011C cDNA, mock-transfected cells (CON), 20 pg of human liver microsomal protein (liver S 5 ) , or 2 pmol of pure P450g (11C13).

S enantiomer). The two IIC18 clones (29c and 6b) represent a second pair

FIGURE 6: Northern blot of RNA from the recombinant transformed COS-I cells. Each lane represents 20 pg of total RNA from mock-transfected cells (control) or cells transfected with the P4501IC clone indicated. The blot was probed with 32P-end-labeledIICSOOR (an oligoprobe which hybridizes with all IIC clones isolated) (top), then stripped, and reprobed with 32P-actincDNA (bottom). A 2.4-kb size marker is indicated at the side of each blot.

of presumed allelic variants, differing by one predicted amino acid change. The 3'-noncoding DNA sequences are similar

Biochemistry, Vol. 30, No. 13, 1991 3255

Cytochrome P450IIC Subfamily

T300R

09 0

m m m m z

m m m z g z E g U l q a

Actin

Northern blot of human mRNAs. Each lane represents I O pg of mRNA, and the blot was probed with end-labeled T300R, an oligoprobe specific for llC8 (top), stripped, and reprobed with 32P-actincDNA (bottom). FIGURE 7:

(a few base differences) for approximately 300 bases, at which point the 2 sequences diverge. At this site, 29c has a polyadenylation signal followed by a poly(A) tail, while clone 6b extends 243 nucleotides further, containing a total of 3 unusual variant polyadenylation signals and no poly(A) tail. These differences could suggest differential splicing, allelic differences, or different genes. At this point, we suggest that 29c and 6b are allelic variants. By the same reasoning, MP-4 and MP-8, which vary by only two amino acids and are now classified as IIC9 and IIClO on the basis of divergent 3’ sequences, may also be only allelic variants of IIC9. I n conclusion, our results indicate there are at least five members of the human P450IIC subfamily. We have examined the profile of cDNAs for this subfamily in a human liver cDNA library from one individual. The predominant members of the IIC subfamily were two allelic variants of IIC9 and IIC8. I1C8 mRNA was low or absent in a second individual phenotypically low in HLx, and no clones for IIC8 were isolated from this individual. Three new members of this subfamily were also identified; IIC17, IIC18, and IIC19. cDNAs for IIC9, IIC18, and IIC19 were expressed in COS-1 cells. Only the two allelic variants of IIC 18 4-hydroxylated racemic mephenytoin at a rate higher than untransfected COS-1 cells, suggesting that IIC18 may be important in mephenytoin metabolism. Additional future studies will address the relative importance of the members of the IIC subfamily in metabolism using other cDNA expression systems. ACKNOWLEDGMENTS We acknowledge Mrs. Patricia Linko for performing the Western blots shown in this paper. We also acknowledge Dr. F. P. Guengerich, Vanderbilt University School of Medicine, Nashville, TN, for generously providing the 4-hydroxymephenytoin standard and the advice of Drs. R. Lindberg and M. Negishi in setting up the COS-1 cell cDNA expression system. REFERENCES Aviv, H., & Leder, P. (1972) Proc. Natl. Acad. Sci. U.S.A. 69, 1408-1412. Beaune, P., Flinois, J.-P., Kiffel, L., Kremers, P., & Leroux,

J.-P. (1985) Biochim. Biophys. Acta 840, 364-370. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254. Brian, W. R., Srivastava, P. K., Umbenhauer, D. R., Lloyd, R. S., & Guengerich, F. P. (1989) Biochemistry 28, 4993-4999. Cox, R. A. (1968) Methods Enzymol. 12, 120-129. Ged, D., Umbenhauer, D. R., Bellew, T. M., Bork, R. W., Srivastava, P. K., Shinriki, N., Lloyd, R. S., & Guengerich, F. P. (1988) Biochemistry 27, 6929-6940. Goldstein, J. A., Linko, P., Luster, M. I., & Sundheimer, D. W. (1982) J . Biol. Chem. 257, 2702-2707. Gut, J., Meier, U. T., Catin, T., & Meyer, U. A. (1986) Biochim. Biophys. Acta 884, 435-447. Kawano, S., Kamataki, T., Yasumori, T., Yamazoe, Y., & Kato, R. (1987) J . Biochem. (Tokyo) 102, 493-501. Kimura, S., Pastewka, J., Gelboin, H. V., & Gonzalez, F. J. (1987) Nucleic Acids Res. 15, 10053-1 0054. Laemmli, U. K. (1970) Nature (London) 227, 680-685. Lasker, J. M., Raucy, J., Kubota, S., Bloswick, B. P., Black, M., & Lieber, C. S. (1987) Biochem. Biophys. Res. Commun. 148, 232-238. Lindberg, R., & Negishi, M. (1989) Nature (London) 339, 6 32-634. Luthman, H., & Magnusson, G. (1983) Nucleic Acids Res. 11, 1295-1308. Meehan, R. R., Gosden, J. R., Rout, D., Rout, N. D., Hastie, N. D., Friedberg, T., Adesnik, M., Buckland, R., van Heyningen, V., Fletcher, J., Spurr, N. K., Sweeney, J., & Wolf, C . R. (1988) Am. J. Hum. Genet. 42, 26-37. Muller, R., Schmidt, W. E., & Stier, A. (1985) FEBS Lett. 187, 21-24. Nebert, D. W., Nelson, D. R., Coon, M. J., Estabrook, R. W., Feyereisen, R., Fujii-Kuriyama, Y., Gonzalez, F. J., Guengerich, F. P., Gunsalus, 1. C., Johnson, E. F., Loper, J. C., Sato, R., Waterman, M. R., & Waxman, D. J. (1991) DNA (in press). Okayama, H., & Berg, P. (1983) Mol. Cell. Biol. 3,280-289. Okino, S . T., Quattrochi, L. C., Pendurthi, U. R., McBride, 0. W., & Tukey, R. H. (1987) J . Biol. Chem. 262, 16072-16079. Relling, M. V., Aoyama, T., Gonzalez, F. J., & Meyer, U. A. (1990) J . Pharmacol. Exp. Ther. 252, 442-447. Romkes, M., Raucy, J. L., & Goldstein, J. A. (1990) FASEB J . 4, A2244. Sanger, F., Coulson, A. R., Hong, G. F., Hill, D. F., & Peterson, G. B. (1982) J . Mol. Biol. 162, 729-773. Shephard, E. A., Phillips, I. R., Santisteban, I., Palmer, C. N. A., & Povey, S. (1989) Ann. Hum. Genet. 53, 23-3 1. Shimada, T., Misono, K. S., & Guengerich, F. P. (1986) J . Biol. Chem. 261, 909-921. Umbenhauer, D. R., Marin, M. V., Lloyd, R. S., & Guengerich, F. P. (1987) Biochemistry 26, 1094-1 099. Watson, C. J., & Jackson, J. F. (1985) in DNA Cloning (Glover, D. M., Ed.) Vol. 1, pp 79-88, IRL Press, Washington, DC. Wrighton, S. A., Thomas, P. E., Willis, P., Maines, S . L., Watkins, P. B., Levin, W., & Guzelian, P. S . (1987) J. Clin. Invest. 80, 1017-1022. Yasumori, T., Kawano, S., Nagata, K., Shimada, M., Yamazoe, y., & Kato, R. (1987) J. Biochem. 102, 1075-1082. Yasumori, T., Murayama, N., Yamazoe, Y., Abe, A., Nogi, Y., Fukasawa, T., & Kato, R. (1989) Mol. Pharmacol. 35, 443-449. Yeowell, H. N., Linko, P., Hodgson, E., & Goldstein, J. A. (1985) Arch. Biochem. Biophys. 243, 408-41 9.